You can run Jev at home with Loki as well! It’s a full agent harness github.com/wundercorp/loki
PatternsEscalation
Human review
The explicit path for uncertain or high-stakes cases: route to a person instead of acting on a low-confidence answer.
- Category
- Patterns
- Also known as
- Escalation
- Related terms
- 4
- Directory entries
- 2
- Docs
- docs.typesafe.ai
- Added
- 2026-09-24
Definition
Every confidence pattern in the docs ends in an escalation path: below a floor, route to a human; between levels, ask the user to confirm; in the entity-alignment cookbook, the middle score level goes to a curator rather than an automatic merge.
Making the path explicit in code is the design point. If the model cannot express honest uncertainty, the system cannot be trusted to run unattended, so the abstain path is part of the architecture rather than an exception handler.
Tagspatternsconfidence
Related terms
Definitions that connect to this one.
Confidence floorA baseline below which code never acts on an answer, regardless of what the answer says.Answers & ConfidenceUncertaintyThe honest "I'm not sure" signal that lets a system defer instead of guessing, exposed through probabilities and confidence.Answers & ConfidenceConfidence-gated routingUse the answer to decide what to do and confidence to decide whether to act, with per-action thresholds set by the cost of being wrong.PatternsUncertainty bandA middle range of probabilities, such as 0.30 to 0.70, that maps to an explicit uncertain outcome instead of a hard yes or no.Answers & Confidence
From the directory
How TypeSafe reports certainty, how it differs from probability, and how to use it architecturally to gate and route decisions.
Practices & PatternsDocs#official#confidence#routing
Official
Use the answer to decide what to do and confidence to decide whether to act: the voice-banking example routes below 0.6 to a human and needs 0.85 or more to auto-approve a transfer.
Practices & PatternsDocs#official#patterns#confidence
Official
From the community
Posts from builders shipping with Jev right now.
Loki: a self-improving harness with a Jev companion
Jev as an LLM guardrail inside Agentgateway
TypeSafe’s new Jev model is fast 🔥 and Agentgateway keeps up! In minutes, I had Jev running as a prompt guard through Agentgateway, with distributed tracing and cost tracking. Check out the example below Example: github.com/agentgateway/a… Intro to Jev: typesafe.ai/blog/introduci…
Jev reranks station suggestions in TrainLCD
提案駅のリランク判定と実測スパイクを追加する by TinyKitten · Pull Request #33 · TrainLCD/Functions github.com/TrainLCD/Funct… AIチャット機能の一部にもJevを潜ませてみるゾ
Vercel's fx safety reviewer, measured at 18× faster
@vercel 's fx uses Jev to review every command. @rauchg says it's up to 18x faster than GPT Luna and more accurate. @JoshARosen built Foreman, where Jev watches coding agents for drift, missed tests and getting stuck. @shantanugoel's Hermes agent installed an "ask Jev" skill on Show more
A game where you race Jev at spotting slop
A small game to show how fast TypeSafe's Jev actually is. Try to beat it at detecting a spam message beatjev2it.jlamberts86.workers.dev
Jev plays itself at 3D chess
got access to jev and instead of doing anything useful i made it play itself at 3D chess same model on both sides, slowly developing beef with itself you can watch jev vs jev or jump in and ruin its day chess-jev.loomens.com
@CompleteSkeptic
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x